CN113159106B - Load curve clustering method, medium and equipment based on morphological trend characteristics - Google Patents
Load curve clustering method, medium and equipment based on morphological trend characteristics Download PDFInfo
- Publication number
- CN113159106B CN113159106B CN202110218252.1A CN202110218252A CN113159106B CN 113159106 B CN113159106 B CN 113159106B CN 202110218252 A CN202110218252 A CN 202110218252A CN 113159106 B CN113159106 B CN 113159106B
- Authority
- CN
- China
- Prior art keywords
- load
- morphological
- trend
- curve
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000000877 morphologic effect Effects 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 60
- 230000005611 electricity Effects 0.000 claims abstract description 78
- 238000005259 measurement Methods 0.000 claims abstract description 43
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000009826 distribution Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims description 56
- 238000003860 storage Methods 0.000 claims description 16
- 231100001263 laboratory chemical safety summary Toxicity 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 6
- 230000004660 morphological change Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000011524 similarity measure Methods 0.000 claims description 3
- 238000013016 damping Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000001174 ascending effect Effects 0.000 description 4
- 238000005452 bending Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010921 in-depth analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a load curve clustering method, medium and equipment based on morphological trend characteristics, and a typical electricity load curve is extracted; extracting morphological characteristics of a typical power load curve based on the variation of load data relative to a mean value, and obtaining trend characteristics through differential processing; measuring morphological feature distance and trend feature distance through the longest public subsequence algorithm, and carrying out multi-scale similarity measurement on a typical power load curve by combining Euclidean distance; and carrying out AP clustering on the user electricity load curves based on the multi-scale curve similarity measurement, classifying the electricity load curves of different users, and dividing users with different electricity modes. The invention can give consideration to the distribution characteristics, morphological characteristics and trend characteristics of the curves, and can more reasonably and effectively distinguish the differences between the curves.
Description
Technical Field
The invention belongs to the technical field of power system analysis, and particularly relates to a load curve clustering method, medium and equipment based on morphological trend characteristics.
Background
With the development of smart grids, a large number of smart meters are used to collect user electricity data, which provides opportunities for more in-depth analysis of electricity behavioral characteristics of individuals and customer groups. The user electricity consumption mode analysis can provide references for electricity price strategy establishment, electricity marketing plan establishment, electricity consumption load prediction and the like, and has great research value.
At present, a partition-based method is often used for clustering the electricity load curves, euclidean distance between the curves is usually calculated to be used as a classifying basis, the method is simple in structure and high in running speed, and the common K-means, fuzzy C-means and other algorithms all have the problems that the number of clusters needs to be set, the dependence of a clustering result on an initial clustering center is high, and local optimal solutions are easy to fall into. Meanwhile, the Euclidean distance is the difference between point-to-point measurement schemes which can only measure curves on the whole, morphology and trend characteristics can not be captured, the morphology and trend characteristics can reflect the difference of power consumption modes of users, and classification of the users is more orderly and reliable; in order to improve the similarity measurement problem, the user classification is carried out by adopting a double-layer structure and combining differential cosine distance and Euclidean distance measurement curve similarity, and a multi-scale similarity measurement mode is proposed by introducing dynamic bending distance and combining an entropy weight method. However, the differential cosine distance is sensitive to noise, and the dynamic bending distance has high calculation time complexity and certain limitation.
Meanwhile, the electricity sampling data generally has certain problems of noise, phase shift and the like, the existing scheme increases the calculation complexity, the algorithm effectiveness is not high, and the problem of load curve clustering cannot be well solved.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a load curve clustering method, medium and equipment based on morphological trend characteristics, which can ignore the numerical magnitude difference existing between curves, reduce noise interference, excavate the similarity between morphological characteristics as much as possible and effectively measure the similarity of the curves; meanwhile, a multi-scale similarity measurement mode taking account of curve distribution characteristics, morphological characteristics and trend change characteristics is constructed, the accuracy of load curve clustering is improved, and meanwhile, the method has higher time efficiency compared with a dynamic bending distance scheme.
The invention adopts the following technical scheme:
a load curve clustering method based on morphological trend characteristics, extracting a typical electricity load curve of a power grid; extracting morphological characteristics of a typical power load curve based on the variation of load data relative to a mean value, and obtaining trend characteristics through differential processing; measuring morphological feature distance and trend feature distance through the longest public subsequence algorithm, and carrying out multi-scale similarity measurement on a typical power load curve by combining Euclidean distance; and carrying out AP clustering on the user electricity load curves based on the multi-scale curve similarity measurement, classifying the electricity load curves of different users, and dividing users with different electricity modes.
Specifically, the typical power load curve is extracted specifically as follows:
selecting a daily load electricity utilization curve of a user in a certain period; removing non-working day and all-zero data, and removing users with missing values exceeding 10% to obtain electricity utilization data of n users on T working days, wherein the electricity utilization curve of the ith user on the T th day is The electricity consumption data of the kth period is obtained, and m is the acquired data number of one day; the average value of the power consumption load of each user in the corresponding period is obtained, and the power consumption load is subjected to range normalization processing to obtain a data set X= { X formed by a typical power consumption load curve 1 ,X 2 ,…,X n } T 。
Specifically, the method for extracting morphological characteristics and trend characteristics specifically comprises the following steps:
calculating X for each typical load curve i Mean value and x value of each time point t it Difference is made with the average value to obtain an average difference matrix D md ＝{d md1 ,d md2 ,…,d mdi …,d mdn } T The method comprises the steps of carrying out a first treatment on the surface of the Characterizing load data relative to negative using quantilesThe change quantity of the load mean value is used for converting the original numerical data into discrete characteristic data uniformly describing the morphological change, the quantile number adopts 3 quantiles, 0.05,0.5,0.95 is respectively taken, and the morphological characteristic x of the ith time point of the ith load curve in the morphological characteristic matrix is calculated mdit The method comprises the steps of carrying out a first treatment on the surface of the For morphological feature matrix X md Performing first-order differential operation to obtain a trend feature matrix X' md 。
Further, the morphological feature x of the ith load curve at the t-th time point in the morphological feature matrix mdit The method comprises the following steps:
wherein d mdit Is the ith mean difference curve d mdi A value at time t.
Specifically, the LCSS algorithm is used to calculate the morphology distance and the trend distance, and the multi-scale similarity measurement mode is constructed by combining the euclidean distance specifically as follows:
calculating a load curve X i And X j The Euclidean distance between the two points is used for measuring the overall distribution characteristics; searching for two-bar state characteristic curve X using LCSS algorithm mdi And X mdj To measure X between load curves i And X j Form similarity between the two images to obtain a form similarity distance D md The method comprises the steps of carrying out a first treatment on the surface of the Extraction of two trend characteristic curves X 'using LCSS algorithm' mdi And X' mdj Calculating trend distance between load curves, measuring local trend characteristics, and calculating trend distance D td 。
Further, the method for constructing the multi-scale similarity measure by combining the morphological distance and the trend distance with the Euclidean distance is specifically as follows: the similarity measurement mode taking the daily load curve distribution characteristics, the morphological characteristics and the local trend characteristics into consideration is constructed and comprises three parts: d (D) all (X i ,X j )＝αD ed (X i ,X j )+βr e,m D md (X i ,X j )+γr e,t D td (X′ i ,X′ j ) Wherein D is all (X i ,X j ) Representing a load curve X i And X j The total distance between the two is respectively the similarity matrix D ed (X i ,X j )、D md (X i ,X j ) D (D) td (X′ i ,X′ j ) Weight coefficient of (c) in the above-mentioned formula (c).
Further, the determining the similarity matrix weight parameters α, β, γ is specifically: setting alpha and gamma to 0 respectively, and finding out the optimal ratio of beta to alpha and gamma respectively according to the step length of 0.1 to combine alpha 0 +β 0 =1 and β 1 +γ 1 Range of fixed beta [ beta ] 0 ,β 1 ]If beta 0 <β 1 Further fine tuning to determine the optimal parameter combinations, using the scaling factor r, respectively e,m And r e,t The uniformity of the value ranges of the three schemes is realized, and the method is specifically calculated as follows:
wherein max (D ed (X i ,X j ))、max(D md (X i ,X j ) And max (D) td (X′ i ,X′ j ) A maximum value of the Euclidean distance of the load curve, a maximum value of the morphological distance and a maximum value of the trend distance respectively.
Specifically, the AP clustering using the multi-scale similarity measure as the algorithm input is specifically:
initializing a similarity matrix S, an attraction matrix R and a attribution matrix A; determining similarity matrix weight parameters alpha, beta and gamma, and calculating a similarity matrix S= [ S (i, j) of n electricity load curves] n×n S (i, j) is the similarity between the ith curve and the jth curve; setting a reference degree p parameter; updating an attraction degree matrix R and a attribution degree matrix A; introduction resistanceThe coefficient lambda maintains the convergence speed and stability of the attribution degree matrix and the attraction degree matrix; if the iteration times exceed the set maximum value or after multiple iterations meet the target, stopping calculation, and determining a clustering center and each cluster; if the number of clusters reachesAnd finally, calculating DB indexes and SSE indexes according to the clustering results to find the optimal results, and finally obtaining z user electricity utilization modes, wherein the electricity utilization mode set is W= [ W ] 1 ,w 2 ,…,w z ]。
Another aspect of the invention is a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform any of the methods.
Another aspect of the present invention is a computing device, including:
one or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing any of the methods.
Compared with the prior art, the invention has at least the following beneficial effects:
the invention provides a load curve clustering method based on morphological trend characteristics, which comprises the steps of firstly, carrying out data preprocessing on original data and removing abnormal values; then extracting morphological characteristics of the load curve, carrying out unified re-expression of the morphological characteristics through quantiles, and simultaneously combining a differential derivative formula to obtain unified trend characteristics; the method comprises the steps of combining an LCSS algorithm to measure the morphology and trend similarity between different curves, adding the distribution characteristics between Euclidean distance measurement curves, and constructing a multi-scale similarity measurement mode taking the distribution characteristics, morphology characteristics and trend characteristics of the curves into consideration; introducing the proposal into an AP clustering algorithm, carrying out cluster analysis on the electricity utilization mode of the user, and extracting the typical electricity utilization mode of the user; based on a mean difference algorithm and a quantile number, a brand new morphological feature extraction and re-expression method is provided, the similarity between LCSS measurement curves is combined, the influence of the magnitude of data is ignored, and the interference of noise is reduced; the similarity measurement scheme taking the local distribution characteristics, the curve morphological characteristics and the curve trend characteristics into consideration is provided, and the overall difference between the curves can be comprehensively focused; the method has advantages in clustering effectiveness and efficiency, and meanwhile, the morphological distance method provided by the invention has higher precision, can identify the morphological characteristics of the curve, and has practical application value.
Furthermore, the purpose of extracting the typical electricity load curve is to primarily screen out abnormal data, reject all zero data and missing data which have smaller meaning on electricity mode analysis and research, and further calculate a typical electricity load curve which can reflect the general characteristics of electricity consumption of users.
Furthermore, the change quantity of the power load curve relative to the curve mean value is extracted through the mean value differential scheme, and the change degree is further measured by the quantile, so that the interference of noise on curve change characteristic extraction can be effectively reduced, and the change trend measurement trend characteristics of the curve between adjacent time points can be reflected by carrying out first-order differential derivative operation on the morphological characteristic matrix on the basis.
Furthermore, by setting the quantile number, the user electricity utilization curve is subjected to unified feature extraction and re-expression, the situation of misclassification caused by the magnitude order difference between the curves can be avoided, the relative change degree of sampling points at each moment of the curve is focused, and the morphological characteristics of the curve can be reflected.
Furthermore, the morphological feature matrix expressed by quantiles and the trend feature matrix subjected to further differential derivative operation all comprise limited discrete feature attribute values, and common subsequences between any morphological feature curves and between any trend feature curves are respectively searched by an LCSS algorithm, so that common features of the load curve, which change in morphological and trend aspects, can be effectively discovered.
Furthermore, a multi-scale similarity measurement mode is constructed based on the morphological distance and the trend distance between the curves and combining with the Euclidean distance, so that measurement of the distribution characteristics of the curves based on the Euclidean distance and measurement of the morphological change characteristics and the trend change characteristics of the curves based on the quantile and the LCSS algorithm can be comprehensively considered, and the method has rationality and effectiveness.
Furthermore, the similarity matrix weight parameters alpha, beta and gamma reflect the contribution degree of different curve similarity measurement modes to the final clustering result, and the weight proportion of different schemes can be reasonably distributed through the control of the weight values of the different curve similarity measurement modes, so that the effective classification of the load curves is realized.
Furthermore, the multi-scale similarity measurement mode is used as the similarity matrix input of the AP clustering algorithm, so that the distribution characteristics, the morphological characteristics and the trend characteristics among the load curves can be considered, the comprehensive consideration of the differences among the curves is realized, and the effective classification is realized.
In conclusion, the method can give consideration to the distribution characteristics, the morphological characteristics and the trend characteristics of the curves, and can more reasonably and effectively distinguish the differences between the curves.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram showing the original classification of UCI data sets, wherein (a) is a standard trend, (b) is a cyclic trend, (c) is an ascending trend, (d) is a descending trend, (e) is a steep ascending trend, and (f) is a steep descending trend;
FIG. 3 is a schematic diagram of a cluster obtained by clustering on a UCI data set according to the clustering method provided by the invention, wherein (a) is a standard trend, (b) is a cyclic trend, (c) is an ascending trend, (d) is a descending trend, (e) is a steep ascending trend, and (f) is a steep descending trend;
FIG. 4 is a schematic diagram of a clustering center obtained after clustering on measured data by the clustering method provided by the invention, wherein (a) is a first type of load center, (b) is a second type of load center, (c) is a third type of load center, and (d) is a fourth type of load center;
fig. 5 is a schematic diagram of a clustering cluster obtained after clustering on measured data by the clustering method provided by the invention, wherein (a) is a first type load, (b) is a second type load, (c) is a third type load, and (d) is a fourth type load.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
It is also to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Various structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the accompanying drawings. The figures are not drawn to scale, wherein certain details are exaggerated for clarity of presentation and may have been omitted. The shapes of the various regions, layers and their relative sizes, positional relationships shown in the drawings are merely exemplary, may in practice deviate due to manufacturing tolerances or technical limitations, and one skilled in the art may additionally design regions/layers having different shapes, sizes, relative positions as actually required.
The invention provides a load curve clustering method based on morphological trend characteristics, which comprises the steps of firstly, carrying out data preprocessing on original data and removing abnormal values; then extracting morphological characteristics of the load curve, carrying out unified re-expression of the morphological characteristics through quantiles, and simultaneously combining a differential derivative formula to obtain unified trend characteristics; the method comprises the steps of combining an LCSS algorithm to measure the morphology and trend similarity between different curves, adding the distribution characteristics between Euclidean distance measurement curves, and constructing a multi-scale similarity measurement mode taking the distribution characteristics, morphology characteristics and trend characteristics of the curves into consideration; the proposal is introduced into an AP clustering algorithm to perform clustering analysis on the user electricity consumption mode, and the typical electricity consumption mode of the user is extracted.
Referring to fig. 1, the load curve clustering method based on morphological trend features of the present invention includes the following steps:
s1, extracting a typical power load curve;
inputting user daily load electricity utilization data in a period of time of a power grid, removing non-working day and all-zero data on the basis, and removing users with the missing values exceeding 10% to obtain electricity utilization data of n users in T working days, wherein the electricity utilization curve of the ith user in the T working days is that Wherein (1)>For the electricity data of the kth period, m is the acquired data number of one day, and the electricity data is acquired once every 30min, so that m is 48; on the basis, the average value of the electricity load of each user at the same moment in the T period is taken as a typical electricity load curve L i ＝{l i1 ,l i2 ,…,l ik ,…,l im A typical power consumption l at time k ik Is calculated as follows:
and to L i Performing range normalization processing to obtain a data set X= { X composed of typical power load curves 1 ,X 2 ,…,X k ,…,X n } T ，X i Normalized electricity data for the ith user, the kth element x thereof ik Is calculated as follows:
wherein max (L i ) And min (L) i ) Respectively typical electrical load curves L i And the maximum and minimum of (a) are defined.
S2, extracting morphological characteristics and trend characteristics of a typical power load curve;
aiming at the normalized user typical power load curve extracted in the step S1, morphological characteristics and trend characteristics are extracted, and the method specifically comprises the following steps:
s201, calculating X of each typical load curve i Mean value and x value of each time point t it Difference is made with the average value to obtain an average difference matrix D md ＝{d md1 ,d md2 ,…,d mdi …,d mdn } T Wherein the ith mean difference curve d mdi The value d at time t mdit Is calculated as follows:
d mdit ＝x it -Mean(X i )
wherein Mean (X i ) Is X i The mean value of the curve;
s202, describing the variation of load data relative to a load mean value by adopting quantiles, converting the original numerical data into discrete characteristic data uniformly describing morphological variation, respectively taking 0.05,0.5,0.95 by adopting 3 quantiles for the quantiles, and obtaining a morphological characteristic matrix X md The ith bar state characteristic curve X mdi ＝{x mdi1 ,x mdi2 ,…,x mdit ,…,x mdim T element x mdit The calculation is as follows:
wherein x is mdit Is the morphological feature of the ith load curve at the t-th time point in the morphological feature matrix, max (D mdi ) And min (D) mdi ) Respectively represent D md Maximum and minimum values, d, of the ith mean difference curve of the matrix mdit The mean difference value of the mean difference curve at the time t is obtained;
s203, for morphological feature matrix X md Performing first-order differential operation to obtain a trend feature matrix X' md For reflecting the consistency of trend changes of adjacent sample points on two load curves, and for the element x 'in the trend changes' mdit The calculation is as follows:
wherein j=1, 2, …, m-1, Δt is the time interval between adjacent points, and m is the morphological characteristic curve X mdi Is a length of (c).
S3, constructing a multi-scale curve similarity measurement mode;
aiming at the morphological characteristics and the trend characteristics of the load curve extracted in the step S2, calculating the morphological distance and the trend distance by using an LCSS algorithm, and constructing a multi-scale similarity measurement mode by combining the Euclidean distance specifically comprises the following steps:
s301, calculating a load curve X i And X j The Euclidean distance between the two is measured to obtain overall distribution characteristics, and the overall distribution characteristics are calculated as follows:
wherein D is ed (X i ,X j ) The Euclidean distance between two curves, m is the load curve X i Length x of (x) t 、y t Respectively are curves X i And X j A value at time t;
s302, searching for a two-bar state characteristic curve X by using LCSS algorithm mdi And X mdj To measure X between load curves i And X j Form similarity between the two images to obtain a form similarity distance D md The calculation is as follows:
D md (X i ,X j )＝len(X mdi )-LCSS(X mdi ,X mdj )
wherein D is md (X i ,X j ) For the load curve X i And X j Morphology similarity distance between the two, len (X mdi ) Is a time sequence curve X mdi Length of (D) morphological distance D md The value of (C) is in the range of [0, len (X) mdi )]The smaller the value, the load curve X i And X j The more similar in morphology; wherein LCSS (X) mdi ,X mdj ) Represents X md Any two curves X in the matrix mdi And X mdj The length of the longest common subsequence between the two is used as the basis for judging the morphological similarity between the curves, and is calculated as follows:
wherein X is mdi And X mdj Respectively representing two time series data curves, len (X mdi ) And len (X) mdj ) Representing the lengths of the two curves;
s303, extracting two trend characteristic curves X 'by using LCSS algorithm' mdi And X' mdj Calculating trend distance between load curves, measuring local trend characteristics, trend distance D td Is calculated as follows:
D td (X′ i ,X′ j )＝len(X′ mdi )-LCSS(X′ mdi ,X′ mdj )
wherein D is td (X′ i ,X′ j ) Is the trend similarity distance between the two load curves, len (X' mdi ) For the length of the time sequence curve, X' mdi The trend change sequence after the first-order difference reflects the local trend characteristics of the load curve;
s304, constructing a similarity measurement mode taking the distribution characteristics, the morphological characteristics and the local trend characteristics of the daily load curve into consideration, wherein the similarity measurement mode comprises three parts:
D all (X i ,X j )＝αD ed (X i ,X j )+βr e,m D md (X i ,X j )+γr e,t D td (X′ i ,X′ j )
wherein D is all (X i ,X j ) Representing a load curve X i And X j The total distance between the two is respectively the similarity matrix D ed (X i ,X j )、D md (X i ,X j ) D (D) td (X′ i ,X′ j ) α+β+γ=1, wherein the determination scheme of α, β, γ is: setting alpha and gamma to 0 respectively, and finding out the optimal ratio of beta to alpha and gamma respectively according to the step length of 0.1 to combine alpha 0 +β 0 =1 and β 1 +γ 1 =1, further fixing the range [ β ] of the setting β 0 ,β 1 ](if beta) 0 <β 1 ) Further fine tuning to determine an optimal combination of parameters; because the range of values of different distance schemes is different, the proportional coefficient r is respectively used e,m And r e,t The uniformity of the value ranges of the three schemes is realized, and the calculation is as follows:
wherein max (D ed (X i ,X j ))、max(D md (X i ,X j ) And max (D) td (X′ i ,X′ j ) A maximum value of the Euclidean distance of the load curve, a maximum value of the morphological distance and a maximum value of the trend distance respectively.
S4, carrying out AP clustering on the users based on a multi-scale similarity measurement mode, and analyzing the power consumption mode of the users;
taking the multi-scale similarity measurement mode obtained in the step S3 as algorithm input to perform AP clustering, and specifically comprises the following steps:
s401, initializing a similarity matrix S, an attraction matrix R and a attribution matrix A;
s402, determining similarity matrix weight parameters alpha, beta and gamma, and further calculating a similarity matrix S= [ S (i, j) of n power utilization load curves] n×n The similarity s (i, j) between the ith and jth load curves is calculated as follows:
s(i,j)＝-D all (X i ,X j )，i≠j
wherein, the larger the similarity value is, the higher the similarity of the two load curves is;
s403, setting a reference degree p (reference) parameter;
s404, updating an attraction degree matrix R and a attribution degree matrix A, and calculating as follows:
wherein k.epsilon.1, 2, …, N, r (i, j) is the load curve X i For X j Is a load curve X i As a cluster center X j The suitability of the sample in (a) (i, j) is the load curve X j For load curve X i Is indicative of the degree of assignment of the load curve X j As a load curve X i Is suitable for the clustering center of the (a);
s405, introducing a damping coefficient lambda, maintaining convergence speed and stability of a attribution degree matrix and an attraction degree matrix, and calculating as follows:
wherein r (i, j) t And a (i, j) t Respectively representing the attraction degree and the attribution degree matrix at the t-th iteration;
s406, if the iteration times exceed the set maximum value or the target is met after a plurality of iterations, stopping calculation, and determining a cluster center and each cluster; otherwise, returning to the step S404, and continuing to calculate;
s407, whether the number of clusters reachesIf yes, ending, calculating DB indexes and SSE indexes according to the clustering result to find the optimal result, and finally obtaining z user electricity utilization modes, wherein the electricity utilization mode set is W= [ W ] 1 ,w 2 ,…,w z ]The method comprises the steps of carrying out a first treatment on the surface of the Not reach->The process returns to step S403 and adjusts the reference degree p to continue the calculation.
In yet another embodiment of the present invention, a terminal device is provided, the terminal device including a processor and a memory, the memory for storing a computer program, the computer program including program instructions, the processor for executing the program instructions stored by the computer storage medium. The processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable gate arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computational core and control core of the terminal adapted to implement one or more instructions, in particular adapted to load and execute one or more instructions to implement a corresponding method flow or a corresponding function; the processor according to the embodiment of the invention can be used for clustering load curves based on morphological trend characteristics, and comprises the following steps: extracting a typical electricity load curve; extracting morphological characteristics of a typical power load curve based on the variation of load data relative to a mean value, and obtaining trend characteristics through differential processing; measuring morphological feature distance and trend feature distance through the longest public subsequence algorithm, and carrying out multi-scale similarity measurement on a typical power load curve by combining Euclidean distance; and carrying out AP clustering on the user electricity load curves based on the multi-scale curve similarity measurement, classifying the electricity load curves of different users to obtain users with different electricity modes, and further carrying out reasonable electricity marketing scheme design and recommendation according to the electricity mode characteristics of the users with different electricity types and the actual electricity characteristics of the users.
In a further embodiment of the present invention, the present invention also provides a storage medium, in particular, a computer readable storage medium (Memory), which is a Memory device in a terminal device, for storing programs and data. It will be appreciated that the computer readable storage medium herein may include both a built-in storage medium in the terminal device and an extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also stored in the memory space are one or more instructions, which may be one or more computer programs (including program code), adapted to be loaded and executed by the processor. The computer readable storage medium herein may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory.
One or more instructions stored in a computer-readable storage medium may be loaded and executed by a processor to implement the respective steps of the above-described embodiments with respect to a morphological trend feature-based load curve clustering method; one or more instructions in a computer-readable storage medium are loaded by a processor and perform the steps of: extracting a typical electricity load curve; extracting morphological characteristics of a typical power load curve based on the variation of load data relative to a mean value, and obtaining trend characteristics through differential processing; measuring morphological feature distance and trend feature distance through the longest public subsequence algorithm, and carrying out multi-scale similarity measurement on a typical power load curve by combining Euclidean distance; and carrying out AP clustering on the user electricity load curve based on the multi-scale curve similarity measurement, analyzing the user electricity mode, classifying the users with different electricity modes, and simultaneously taking the historical electricity data of the users as input to predict the future electricity consumption condition of the users.
The practical applicability and effect of the invention were evaluated as follows using the common data set Synthetic Control Chart Time Series (SCCTS) provided by UCI and irish grid measured data to simulate the example:
the clustering effect of the algorithm is compared on the SCCTS data set, and the data comprises: 6 kinds of curves such as standard trend, cyclic trend, rising trend, descending trend, steep rising trend and steep descending trend, and the total of 600 curves are 100 in each kind of curve.
The original classification of 600 curves is shown in fig. 2, and the distribution of each cluster curve after the clustering method based on morphological trend features is adopted is shown in fig. 3.
From the clustering results, it can be seen that the method can accurately distinguish the differences of different curve forms and trends.
Meanwhile, the text algorithm is compared with the algorithm based on Euclidean distance, based on difference cosine distance, based on morphological distance, based on dynamic bending distance and the like from the aspects of clustering effectiveness index, AP clustering algorithm similarity matrix calculation time (SCT) and the like. By comparing the validity of the algorithm with the standard classification results of the dataset.
Each algorithm adjusts the reference degree p through multiple experiments, and the clustering number is 6, and the result is shown in table 1.
Table 1 comparison of clustering indicators of various similarity measurement algorithms
From the aspect of clustering effectiveness indexes, the morphological similarity measurement scheme provided by the invention is obviously superior to the traditional Euclidean distance and difference cosine distance scheme in terms of AR indexes and FM indexes, and is completely acceptable in terms of time efficiency although the clustering effect is lower than that of a DTW algorithm. Meanwhile, the multi-scale similarity measurement mode is superior to the DTW algorithm in terms of AR index and FM index, and the calculation time is lower. Meanwhile, on the DB index, the method is higher than the Euclidean distance and difference cosine distance scheme, but is closer to the DB index of the standard data set, so that the effectiveness of the method in operation time and clustering result is shown.
Fig. 4 and fig. 5 are respectively the results of clustering on the irish practical electricity data set, and it can be found that the obtained 4 types of curves have larger morphological differences and obvious morphological characteristics, the first type of load is a typical unimodal load curve, and is commonly used for office electricity, industrial electricity and the like, the electricity consumption is increased sharply after 8 points, the electricity consumption is reduced sharply after 18 points, and the morphological characteristics are obvious; the second type and the third type of loads are typical resident electricity loads, the second type of loads use basic electricity demand in daytime and the electricity consumption quantity of the second type of loads reaches 18 points to be increased suddenly, and the second type of loads belong to the condition that office workers exist in the home; the third type load reaches an electricity peak value at 21 pm, and has basic electricity demand in daytime, which is a multi-port condition; the fourth type load meets the requirements all the day, and meanwhile, 14 points and 17 points are high in electricity consumption, which is probably the situation of commercial electricity consumption users; therefore, the load curve clustering method of the multi-scale similarity provided by the invention has reasonable results.
In summary, the method, medium and equipment for clustering the load curve based on the morphological trend feature have certain advantages in clustering effectiveness and time efficiency compared with the traditional scheme, and the clustering result on the actual measurement data set is reasonable and effective. .
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above is only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited by this, and any modification made on the basis of the technical scheme according to the technical idea of the present invention falls within the protection scope of the claims of the present invention.
Claims (7)
1. A load curve clustering method based on morphological trend features is characterized by extracting a typical power consumption load curve of a power grid; extracting morphological characteristics of a typical power load curve based on the variation of load data relative to a mean value, and obtaining trend characteristics through differential processing; measuring morphological feature distance and trend feature distance through the longest public subsequence algorithm, and carrying out multi-scale similarity measurement on a typical power load curve by combining Euclidean distance; performing AP clustering on the user electricity load curves based on the multi-scale curve similarity measurement, classifying the electricity load curves of different users, and dividing users with different electricity modes;
the method for constructing the multi-scale similarity measurement mode by combining the Euclidean distance comprises the following steps of:
calculating a load curve X i And X j The Euclidean distance between the two points is used for measuring the overall distribution characteristics; searching for two-bar state characteristic curve X using LCSS algorithm mdi And X mdj To measure X between load curves i And X j Morphological similarity betweenThe degree and then the morphological similarity distance D is obtained md The method comprises the steps of carrying out a first treatment on the surface of the Extraction of two trend characteristic curves X 'using LCSS algorithm' mdi And X' mdj Calculating trend distance between load curves, measuring local trend characteristics, and calculating trend distance D td ；
The method for constructing the multi-scale similarity measurement mode by combining the morphological distance and the trend distance with the Euclidean distance comprises the following specific steps: the similarity measurement mode taking the daily load curve distribution characteristics, the morphological characteristics and the local trend characteristics into consideration is constructed and comprises three parts: d (D) ali (X i ，X j )＝αD ed (X i ，X j )+βr e，m D md (X i ，X j )+γr e，t D td (X′ i ，X′ j ) Wherein D is all (X i ，X j ) Representing a load curve X i And X j The total distance between the two is respectively the similarity matrix D ed (X i ，X j )、D md (X i ，X j ) D (D) td (X′ i ，X′ j ) The weight coefficients of the similarity matrix weight parameters alpha, beta and gamma are determined specifically as follows: setting alpha and gamma to 0 respectively, and finding out the optimal ratio of beta to alpha and gamma respectively according to the step length of 0.1 to combine alpha 0 +β 0 =1 and β 1 +γ 1 Range of fixed beta [ beta ] 0 ，β 1 ]If beta 0 ＜β 1 Further fine tuning to determine the optimal parameter combinations, using the scaling factor r, respectively e，m And r e，t The uniformity of the value ranges of the three schemes is realized, and the method is specifically calculated as follows:
wherein max (D ed (X i ，X j ))、max(D md (X i ，X j ) And max (D) td (X′ i ，X′ j ) A maximum value of the Euclidean distance of the load curve, a maximum value of the morphological distance and a maximum value of the trend distance respectively.
2. The method according to claim 1, characterized in that the extraction of the typical electrical load curve is in particular:
selecting a daily load electricity utilization curve of a user in a certain period; removing non-working day and all-zero data, and removing users with missing values exceeding 10% to obtain electricity utilization data of n users on T working days, wherein the electricity utilization curve of the ith user on the T th day is The electricity consumption data of the kth period is obtained, and m is the acquired data number of one day; the average value of the power consumption load of each user in the corresponding period is obtained, and the power consumption load is subjected to range normalization processing to obtain a data set X= { X formed by a typical power consumption load curve 1 ，X 2 ，…，X n } T 。
3. The method according to claim 1, wherein the extracting morphological features and trend features is specifically:
calculate each typical load curve X i And the value x of each time point t it Difference is made with the average value to obtain an average difference matrix D md ＝{d md1 ，d md2 ，…，d mdi …，d mdn } T The method comprises the steps of carrying out a first treatment on the surface of the Negative sign by quantileThe change quantity of the load data relative to the load mean value is used for converting the original numerical data into discrete characteristic data which uniformly describes the morphological change, the quantile adopts 3 quantiles, 0.05,0.5,0.95 is respectively taken, and the morphological characteristic x of the ith time point of the ith load curve in the morphological characteristic matrix is calculated mdit The method comprises the steps of carrying out a first treatment on the surface of the For morphological feature matrix X md Performing first-order differential operation to obtain a trend feature matrix X' md 。
4. A method according to claim 3, wherein the morphology feature x of the ith time point of the ith load curve in the morphology feature matrix mdit The method comprises the following steps:
wherein d mdit Is the ith mean difference curve d mdi A value at time t.
5. The method according to claim 1, wherein AP clustering using a multi-scale similarity measure as an algorithm input is specifically:
initializing a similarity matrix S, an attraction matrix R and a attribution matrix A; determining similarity matrix weight parameters alpha, beta and gamma, and calculating a similarity matrix S= [ S (i, j) of n electricity load curves] n×n S (i, j) is the similarity between the ith curve and the jth curve; setting a reference degree p; updating an attraction degree matrix R and a attribution degree matrix A; introducing a damping coefficient lambda to maintain the convergence speed and stability of the attribution degree matrix and the attraction degree matrix; if the iteration times exceed the set maximum value or after multiple iterations meet the target, stopping calculation, and determining a clustering center and each cluster; if the number of clusters reachesAnd finally, calculating DB indexes and SSE indexes according to the clustering results to find the optimal results, and finally obtaining z user electricity utilization modes, wherein the electricity utilization mode set is W= [ W ] 1 ，w 2 ，…，w z ]。
6. A computer readable storage medium storing one or more programs, wherein the one or more programs comprise instructions, which when executed by a computing device, cause the computing device to perform any of the methods of claims 1-5.
7. A computing device, comprising:
one or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing any of the methods of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110218252.1A CN113159106B (en) | 2021-02-26 | 2021-02-26 | Load curve clustering method, medium and equipment based on morphological trend characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110218252.1A CN113159106B (en) | 2021-02-26 | 2021-02-26 | Load curve clustering method, medium and equipment based on morphological trend characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113159106A CN113159106A (en) | 2021-07-23 |
| CN113159106B true CN113159106B (en) | 2024-02-02 |
Family
ID=76883534
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110218252.1A Active CN113159106B (en) | 2021-02-26 | 2021-02-26 | Load curve clustering method, medium and equipment based on morphological trend characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113159106B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780440A (en) * | 2021-09-15 | 2021-12-10 | 江苏方天电力技术有限公司 | Low-voltage station area phase identification method for improving data disturbance resistance |
| CN117076990B (en) * | 2023-10-13 | 2024-02-27 | 国网浙江省电力有限公司 | Load curve identification method, device and medium based on curve dimension reduction and clustering |
| CN117932445B (en) * | 2024-03-25 | 2024-05-31 | 西安航科创星电子科技有限公司 | High-stability HTCC alumina ceramic preparation parameter anomaly identification method |
| CN117992856B (en) * | 2024-04-03 | 2024-06-21 | 国网山东省电力公司营销服务中心(计量中心) | User electricity behavior analysis method, system, device, medium and program product |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002169613A (en) * | 2000-12-04 | 2002-06-14 | Hitachi Ltd | Analysis method for electric power load curve and system |
| JP2005245147A (en) * | 2004-02-27 | 2005-09-08 | Mitsubishi Electric Corp | Power information system |
| CN107423769A (en) * | 2017-08-03 | 2017-12-01 | 四川大学 | Electric load curve adaptive clustering scheme based on morphological feature |
| CN107633661A (en) * | 2017-08-28 | 2018-01-26 | 国家电网公司 | Pump-storage generator runout alarm method and device |
| CN109902953A (en) * | 2019-02-27 | 2019-06-18 | 华北电力大学 | A kind of classification of power customers method based on adaptive population cluster |
| WO2020063689A1 (en) * | 2018-09-25 | 2020-04-02 | 新智数字科技有限公司 | Method and device for predicting thermal load of electrical system |
| CN111199361A (en) * | 2020-01-13 | 2020-05-26 | 国网福建省电力有限公司信息通信分公司 | Electric power information system health assessment method and system based on fuzzy reasoning theory |
| CN111832796A (en) * | 2020-02-29 | 2020-10-27 | 上海电力大学 | Fine classification and prediction method and system for residential electricity load mode |
| CN111914287A (en) * | 2020-06-17 | 2020-11-10 | 西安交通大学 | Improved DTW (delay tolerant W) measurement method for track privacy protection, storage device and equipment |
| CN112330028A (en) * | 2020-11-08 | 2021-02-05 | 国网天津市电力公司 | Electric bus charging load prediction method based on spectral clustering and LSTM neural network |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6022050B2 (en) * | 2013-05-16 | 2016-11-09 | 三菱電機株式会社 | Consumer power control system and consumer power control method |
| US9585632B2 (en) * | 2014-04-23 | 2017-03-07 | Siemens Medical Solutions Usa, Inc. | Estimation of a mechanical property of anatomy from medical scan data |
| US10126134B2 (en) * | 2015-12-21 | 2018-11-13 | Invensense, Inc. | Method and system for estimating uncertainty for offline map information aided enhanced portable navigation |
-
2021
- 2021-02-26 CN CN202110218252.1A patent/CN113159106B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002169613A (en) * | 2000-12-04 | 2002-06-14 | Hitachi Ltd | Analysis method for electric power load curve and system |
| JP2005245147A (en) * | 2004-02-27 | 2005-09-08 | Mitsubishi Electric Corp | Power information system |
| CN107423769A (en) * | 2017-08-03 | 2017-12-01 | 四川大学 | Electric load curve adaptive clustering scheme based on morphological feature |
| CN107633661A (en) * | 2017-08-28 | 2018-01-26 | 国家电网公司 | Pump-storage generator runout alarm method and device |
| WO2020063689A1 (en) * | 2018-09-25 | 2020-04-02 | 新智数字科技有限公司 | Method and device for predicting thermal load of electrical system |
| CN109902953A (en) * | 2019-02-27 | 2019-06-18 | 华北电力大学 | A kind of classification of power customers method based on adaptive population cluster |
| CN111199361A (en) * | 2020-01-13 | 2020-05-26 | 国网福建省电力有限公司信息通信分公司 | Electric power information system health assessment method and system based on fuzzy reasoning theory |
| CN111832796A (en) * | 2020-02-29 | 2020-10-27 | 上海电力大学 | Fine classification and prediction method and system for residential electricity load mode |
| CN111914287A (en) * | 2020-06-17 | 2020-11-10 | 西安交通大学 | Improved DTW (delay tolerant W) measurement method for track privacy protection, storage device and equipment |
| CN112330028A (en) * | 2020-11-08 | 2021-02-05 | 国网天津市电力公司 | Electric bus charging load prediction method based on spectral clustering and LSTM neural network |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
Non-Patent Citations (7)
| Title |
|---|
| Galaxy morphological classification in deep-wide surveys via unsupervised machine learning;Monthly Notices of the Royal Astronomical Society;第491卷(第1期);全文 * |
| Pēteris Grabusts ; Arkady Borisov.Clustering Methodology for Time Series Mining.Scientific Journal of Riga Technical University. Construction Science.2009,(第40期),全文. * |
| Scelidosaurus harrisonii from the Early Jurassic of Dorset, England: postcranial skeleton;Zoological Journal of the Linnean Society;第189卷(第1期);全文 * |
| Zhang Zhang ; Kaiqi Huang ; Tieniu Tan.Comparison of Similarity Measures for Trajectory Clustering in Outdoor Surveillance Scenes.万方外文会议文献数据库.2006,全文. * |
| Zhenjun Ma ; Rui Yan ; Natasa Nord.A variation focused cluster analysis strategy to identify typical daily heating load profiles of higher education buildings.Energy.2017,第134卷全文. * |
| 基于形态相似距离的时间序列相似度计算;李中;刘洋洋;张铁峰;;计算机工程与设计(第03期);全文 * |
| 基于曲线形态特征的地区规模化风电出力场景划分;林俐;肖舒;费宏运;潘险险;;电网与清洁能源(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113159106A (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113159106B (en) | Load curve clustering method, medium and equipment based on morphological trend characteristics | |
| Rajabi et al. | A comparative study of clustering techniques for electrical load pattern segmentation | |
| CN111525587B (en) | Reactive load situation-based power grid reactive voltage control method and system | |
| CN112819299A (en) | Differential K-means load clustering method based on center optimization | |
| CN113744089B (en) | Transformer area household variable relation identification method and device | |
| CN110490369A (en) | A kind of Short-Term Load Forecasting Method based on EWT and LSSVM model | |
| Zhang et al. | Short-term load forecasting method based on EWT and IDBSCAN | |
| CN112907062A (en) | Power grid electric quantity prediction method, device, medium and terminal integrating temperature characteristics | |
| CN109544029A (en) | Analysis method, analytical equipment and the terminal of a kind of area's line loss | |
| Shamim et al. | Multi-domain feature extraction for improved clustering of smart meter data | |
| CN115375014A (en) | Source-load combination probability prediction method, device and storage medium | |
| CN112215398A (en) | Power consumer load prediction model establishing method, device, equipment and storage medium | |
| CN112464059B (en) | Distribution network user classification method, device, computer equipment and storage medium | |
| CN117787572A (en) | Abnormal electricity utilization user identification method and device, storage medium and electronic equipment | |
| CN117194872A (en) | Wind power plant wind speed correction method, system, computer equipment and storage medium | |
| CN116862137A (en) | Charging pile load flexible scheduling method and device based on data fusion | |
| CN116565851A (en) | Wind farm power prediction method, device, equipment and medium based on clustering algorithm | |
| Wang et al. | Application of clustering technique to electricity customer classification for load forecasting | |
| CN114723147A (en) | New energy power prediction method based on improved wavelet transform and neural network | |
| Liang et al. | Forecasting tourist arrivals using dual decomposition strategy and an improved fuzzy time series method | |
| CN110175705B (en) | Load prediction method and memory and system comprising same | |
| Wang et al. | Analysis of user’s power consumption behavior based on k-means | |
| Shamim et al. | Novel technique for feature computation and clustering of smart meter data | |
| Wen et al. | High-resolution load profile clustering approach based on dynamic largest triangle three buckets and multiscale dynamic warping path under limited warping path length | |
| CN110516867A (en) | A kind of integrated study load forecasting method based on principal component analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220322 Address after: 730050 No.8, Beibinhe East Road, Chengguan District, Lanzhou City, Gansu Province Applicant after: STATE GRID GANSU ELECTRIC POWER Co. Applicant after: INFORMATION COMMUNICATION COMPANY OF STATE GRID GANSU ELECTRIC POWER Co. Address before: 730046 No.8, Beibinhe East Road, Chengguan District, Lanzhou City, Gansu Province Applicant before: STATE GRID GANSU ELECTRIC POWER Co. Applicant before: XI'AN JIAOTONG University |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |